Understanding Ammonia Emissions from Swine Animal Feeding Operations in North Carolina

Swine animal feeding operations (AFOs) in eastern North Carolina are a source of emissions for ammonia (NH3), hydrogen sulfide (H 2S), carbon dioxide (CO2), volatile organic carbons (VOCs), and fine particulate matter (PM). The majority of observations regarding emissions from swine AFO's in NC have focused on lagoons and not animal housing units. The objectives of this research were: (i) to assess the appropriate value for the NH3 Henry's Law coefficient when modeling NH 3 emission from liquid swine waste; (ii) to review and synthesize ancillary data collected as part of the National Air Emissions Monitoring Study (NAEMS) to better understand nutrient flows through swine housing units; and (iii) to model NH3 emissions from shallow pits in swine housing units using the US EPA WATER9 model. Together these results further the development of process-based models to describe gaseous emissions from swine animal housing units in NC.;Use of the Henry's law coefficient for NH3 is an important part of most models describing NH3 emissions from liquid surfaces. Swine waste that accumulates in the shallow pits under swine housing units in NC, however, is a complicated mixture/solution of water, feces, urine, dissolved organic matter, mineral solids, and microorganisms. Ammonia emissions from three different sources [pure solutions of (NH4)2SO 4 (AS), swine anaerobic lagoon liquid (LL), and pit liquid (PL) from swine housing units] were evaluated using a small flow-through teflon-lined chamber (SFTC; 0.3m x 0.2m x 0.15m) under controlled laboratory conditions. The SFTC was designed for 100% collection efficiency of all NH 3 gas emitted from the liquids. NH3 flux varied linearly with TAN (mg NH3+NH4-N L-1) concentration, but emissions from LL and PL were ∼5x those from the pure (NH 4)2SO4 solutions (Y = 0.02X + 2.3 for AS vs. Y = 0.11X - 2.9 and Y = 0.10X - 4.7 for LL and PL, respectively, Y = mg NH3-N m-2 hr-1, X = TAN; pH 7.6; T = 20°C). Furthermore, the enhancement in NH3 emissions was a function of TAN concentration, decreasing in intensity at higher TAN and approaching rates comparable to the pure solutions of AS. Possible mechanisms that could enhance NH3 emissions from LL and PL were investigated involving CO2, H2S, and VOCs. Only additions of NaHCO3 were found to enhance NH3 emissions from pure AS solutions in a pattern similar to that observed for LL and PL. These results suggest using Henry's Law coefficients derived from simple aqueous solutions substantially underestimate NH3 emissions from LL and PL, and that there may be several mechanisms controlling emissions of NH3 across the air-liquid interface for LL and PL.;Ancillary information regarding animal inventory, feed consumption, growth rates of hogs, manure excretion and water usage gathered by the National Air Emissions Monitoring Study (NAEMS) project were reviewed and synthesized to better understand N inputs and outputs from swine finisher houses using shallow manure pits. The average amount of feed consumed during a rotation was 181,200 kg barn-1 (n=3 barns per rotation; total of 4 rotations; ∼730 hogs per barn per rotation). Up to five different feed formulations were fed during a rotation with %N contents ranging from 3.4 to 2.2%N. Average N consumed per rotation per barn was 4970 +/-500 kg, as determined from farm records (tonnage delivered and delivery dates) and weekly analysis of feed samples from the barns. Assuming an average rotation length of 140 days, ∼35.5 kg N day-1 was consumed by the hogs in the barns (∼50 grams N per hog per day per rotation). Average net live weight gain per barn per rotation was 74000 kg, yielding a feed conversion ratio of 2.4 (kg of feed to produce 1 kg of NWG). Based on total live weight of hogs shipped, ∼2030 kg N was exported per barn per rotation as animal mass. The remaining N introduced via feed was excreted as urine and fecal matter (2940 kg N barn-1 rotation-1; ∼29 grams N hog-1 day-1). TAN content and pH of pit liquid was consistently higher (∼1885 +/-389.27 mg TAN L-1; pH 7.7) than that found in the anaerobic lagoon (802 +/-72.78 mg TAN L-1; pH 7.9).;Sources of NH3 emissions in swine housing units with shallow pits are restricted to the air/liquid interface of the pit liquid, and urine and fecal matter that remain on the flooring within each pen. Emissions of NH3 from the pit liquid were modeled for periods of time during a specific rotation using the WATER9. Modeled results for NH3 emissions were compared to measured emissions from the housing units for the same time periods. To date, WATER9 over predicts NH3 emissions by a factor of 1.2 as compared measured emissions by the NAEMS project, limiting its use to estimate emissions from the flooring by difference.